Research On Collaborative Task Planning Of Multiple Fire Robots In Fire Environment

With the development of robot industrialization and intelligence,more and more fire robots are replacing firefighters to join the firefighting and rescue work.In order to improve the fire disposal ability of multiple fire robots,it is crucial to use technologies such as collaborative task allocation,path planning,and formation control of multiple fire robots.The following research is conducted:To optimize the task allocation problem of firefighting robots,an improved clustering CBBA(Consensius-Based Bundle Algorithm)algorithm is designed to solve it.Firstly,establish a mathematical model for collaborative firefighting and rescue of multiple firefighting robots.Secondly,the distance reward and punishment coefficient and clustering function are introduced to improve the CBBA algorithm.Finally,the improved clustering CBBA algorithm is applied to solve the task allocation problem of fire robots.The results show that the improved clustering CBBA algorithm reduces the length of the robot’s travel path and improves the safety and efficiency of the robot.To improve the obstacle avoidance and optimization capabilities of fire robot path planning,an improved ant colony algorithm is proposed and optimized to find a collision free and optimal smooth path.Firstly,establish a static environmental map model using the grid method.Secondly,traditional ant colony algorithms are improved by introducing initial grid transfer rules,establishing ant colony communication mechanisms,improving path smoothness and security reliability.Finally,conduct simulation verification.Compared with the original algorithm,the improved ant colony algorithm has higher path security,fewer turns,and significantly reduced pathfinding time.In order to maintain the formation of fire robots,the leader follower group control problem of multiple fire robots based on distributed estimation was studied.Firstly,a trajectory tracking controller was designed for the leader firefighting robot to follow the reference trajectory;Secondly,a distributed estimator was designed to enable each following fire robot to estimate the state of the leading fire robot.Then,a distributed control strategy based on biological neural dynamics was proposed to solve the speed jump problem.Finally,the effectiveness of the proposed estimator and controller is verified through simulation.The research in this article broadens the basic theory of task allocation and path planning for fire robots,enriches the algorithm research of robot problems,and has certain theoretical value and practical significance for the future development of fire robots.